CN115115620A - Pneumonia lesion simulation method and system based on deep learning - Google Patents
Pneumonia lesion simulation method and system based on deep learning Download PDFInfo
- Publication number
- CN115115620A CN115115620A CN202211012475.3A CN202211012475A CN115115620A CN 115115620 A CN115115620 A CN 115115620A CN 202211012475 A CN202211012475 A CN 202211012475A CN 115115620 A CN115115620 A CN 115115620A
- Authority
- CN
- China
- Prior art keywords
- image
- lesion
- simulation
- pneumonia
- lung
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30061—Lung
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
- G06V2201/031—Recognition of patterns in medical or anatomical images of internal organs
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a pneumonia lesion simulation method and system based on deep learning. The method comprises the steps of obtaining basic CT lung images of a patient, carrying out health image simulation processing on the lung images to obtain initial health simulation lung images, obtaining historical lung pathological change image sets from a lung pneumonopathy image database by constructing the lung pneumonopathy image database, and leading the image sets into a deep learning-based pneumonia change model for image training to obtain pneumonia pathological change process parameters. And importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model to perform lesion simulation to obtain a lesion simulation image set, and performing data comparison and analysis on the patient real CT lung image and the lesion simulation image set to obtain pneumonia condition reference information so as to assist a doctor in judging the pneumonia condition of the patient. The invention can assist doctors in comprehensively analyzing the lung images of patients, improve the working efficiency of doctors and reduce the misdiagnosis rate.
Description
Technical Field
The invention relates to the field of deep learning, in particular to a pneumonia lesion simulation method and system based on deep learning.
Background
The incidence of pneumonia is high, accounting for about 12% of the total population. 250 million community-acquired pneumonia (CAP) patients are expected to be present in China each year, with over 12 million people dying of community-acquired pneumonia. In addition, the incidence of pneumonia tends to increase in recent years due to aging of social population, increase in immunocompromised hosts, pathogen migration, difficulty in etiological diagnosis, increase in bacterial drug resistance caused by improper use of antibiotics, and the like.
With increasing incidence, pneumonia is seen more and more. When pneumonia is checked, if a doctor only checks CT lung images of a patient, the process is time-consuming and labor-consuming, and under the condition that a large number of patients exist, doctors are also greatly examined, so that a method capable of assisting the doctors to make lesion judgment on the CT images of the pneumonia patients is needed, and the diagnosis efficiency of the doctors is improved.
Disclosure of Invention
In order to solve at least one technical problem, the invention provides a pneumonia lesion simulation method and system based on deep learning.
The invention provides a pneumonia lesion simulation method based on deep learning in a first aspect, which comprises the following steps:
acquiring an initial CT lung image, and performing image preprocessing on the initial CT lung image to obtain a basic lung image;
building a lung health simulation model based on deep learning, and performing image key data extraction and image data simulation on a basic lung image to obtain an initial health simulation lung image;
importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model for lesion simulation to obtain a lesion simulation image set;
and collecting the lesion simulation image into an image for splitting, and displaying in a preset terminal device according to a preset mode.
In the scheme, the acquiring of the initial CT lung image and the image preprocessing of the initial CT lung image are to obtain a basic lung image, which specifically comprises:
acquiring an initial CT lung image of a target object, and performing data format conversion processing on the initial CT lung image to obtain a lung image with a uniform format;
and carrying out image smoothing and noise reduction on the lung images with the uniform format to obtain high-quality basic lung images.
In the scheme, the lung health simulation model based on deep learning is built, and the basic lung image is subjected to image key data extraction and image data simulation to obtain an initial health simulation lung image, which specifically comprises the following steps: carrying out image contour sharpening on the basic lung image to obtain a sharpened basic lung image;
carrying out image contour feature extraction on the sharpened basic lung image to obtain lung size information and contour information;
and carrying out image data simulation analysis on the lung size and contour information, and generating an initial health simulation lung image.
In this scheme, the initial healthy simulated lung image and pneumonia pathological change process parameters are imported into a pneumonia change model for pathological change simulation, so as to obtain a pathological change simulated image set, and the method comprises the following steps:
acquiring an existing historical pneumonia lesion image set from a pneumonia lesion image database;
splitting image data of the historical pneumonia lesion image set according to a preset sequence to obtain a plurality of pneumonia lesion image subsets;
and importing the plurality of pneumonia lesion image subsets into a pneumonia change model for image change analysis, and obtaining pneumonia lesion process parameters according to image changes among different pneumonia lesion image subsets.
In this scheme, carry out the pathological change simulation in importing the initial healthy simulation lung image and pneumonia pathological change process parameter into pneumonia change model, obtain pathological change simulation image set, still include:
acquiring basic condition information of a target object, and performing data analysis processing on the basic condition information of the target object to obtain basic data of the target object;
performing pneumonia lesion influence analysis on the target object basic data to obtain a plurality of lesion influence indexes, and calculating an average lesion influence index;
and analyzing to obtain the pneumonia lesion process parameter correction information according to the average lesion influence index.
In this scheme, carry out the pathological change simulation in importing the initial healthy simulation lung image and pneumonia pathological change process parameter into pneumonia change model, obtain pathological change simulation image set, still include:
correcting the pneumonia lesion process parameters according to the lesion process parameter correction information;
importing the initial health simulation lung image and the corrected pneumonia lesion process parameters into a pneumonia change model for lesion process simulation to obtain lesion simulation image data;
dividing the lesion simulation image data into image data according to the image change degree to obtain a plurality of lesion simulation image subsets;
and performing data sorting on the plurality of lesion simulation image subsets to obtain a lesion simulation image set.
The second aspect of the present invention also provides a pneumonia lesion simulation system based on deep learning, including: the pneumonia lesion simulation method program based on deep learning is executed by the processor, and the pneumonia lesion simulation method program based on deep learning realizes the following steps:
acquiring an initial CT lung image, and performing image preprocessing on the initial CT lung image to obtain a basic lung image;
building a lung health simulation model based on deep learning, and performing image key data extraction and image data simulation on a basic lung image to obtain an initial health simulation lung image;
importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model for lesion simulation to obtain a lesion simulation image set;
and collecting the lesion simulation image into an image for splitting, and displaying in a preset terminal device according to a preset mode.
In the scheme, the acquiring of the initial CT lung image and the image preprocessing of the initial CT lung image are to obtain a basic lung image, which specifically comprises:
acquiring an initial CT lung image of a target object, and performing data format conversion processing on the initial CT lung image to obtain a lung image with a uniform format;
and carrying out image smoothing and noise reduction on the lung images with the uniform format to obtain high-quality basic lung images.
In this scheme, carry out the pathological change simulation in importing the initial healthy simulation lung image and pneumonia pathological change process parameter into pneumonia change model, obtain pathological change simulation image set, still include:
acquiring basic condition information of a target object, and performing data analysis processing on the basic condition information of the target object to obtain basic data of the target object;
performing pneumonia lesion influence analysis on the target object basic data to obtain a plurality of lesion influence indexes, and calculating an average lesion influence index;
and analyzing to obtain the pneumonia lesion process parameter correction information according to the average lesion influence index.
In this scheme, the method of importing the initial healthy simulated lung image and the pneumonia pathological change process parameters into a pneumonia change model for pathological change simulation to obtain a pathological change simulated image set further comprises:
correcting the pneumonia lesion process parameters according to the lesion process parameter correction information;
importing the initial health simulation lung image and the corrected pneumonia lesion process parameters into a pneumonia change model for lesion process simulation to obtain lesion simulation image data;
dividing the lesion simulation image data into image data according to the image change degree to obtain a plurality of lesion simulation image subsets;
and performing data sorting on the plurality of lesion simulation image subsets to obtain a lesion simulation image set.
The invention discloses a pneumonia lesion simulation method and system based on deep learning. The method comprises the steps of obtaining basic CT lung images of a patient, carrying out health image simulation processing on the lung images to obtain initial health simulation lung images, obtaining historical lung pathological change image sets from a lung pneumonopathy image database by constructing the lung pneumonopathy image database, and leading the image sets into a deep learning-based pneumonia change model for image training to obtain pneumonia pathological change process parameters. And importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model to perform lesion simulation to obtain a lesion simulation image set, and performing data comparison and analysis on the patient real CT lung image and the lesion simulation image set to obtain pneumonia condition reference information so as to assist a doctor in judging the pneumonia condition of the patient. The invention can assist doctors in comprehensively analyzing the lung images of patients, improve the working efficiency of doctors and reduce the misdiagnosis rate.
Drawings
FIG. 1 is a flow chart of a deep learning-based pneumonia lesion simulation method according to the present invention;
FIG. 2 illustrates a flow chart of an initial health simulation lung image acquisition of the present invention;
FIG. 3 illustrates a lesion simulation image set acquisition flow chart of the present invention;
fig. 4 shows a block diagram of a pneumonia lesion simulation system based on deep learning according to the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
Fig. 1 shows a flowchart of a pneumonia lesion simulation method based on deep learning according to the present invention.
As shown in fig. 1, a first aspect of the present invention provides a pneumonia lesion simulation method based on deep learning, including:
s102, acquiring an initial CT lung image, and performing image preprocessing on the initial CT lung image to obtain a basic lung image;
s104, building a lung health simulation model based on deep learning, and performing image key data extraction and image data simulation on a basic lung image to obtain an initial health simulation lung image;
s106, importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model for lesion simulation to obtain a lesion simulation image set;
and S108, collecting the lesion simulation image into an image for splitting, and displaying in a preset terminal device according to a preset mode.
It should be noted that the preset mode is to send the images in the lesion simulation image set to a preset terminal device in a time sequence, where the preset terminal device includes a mobile terminal device and a computer terminal device.
According to the embodiment of the invention, the acquiring of the initial CT lung image and the image preprocessing of the initial CT lung image are used for obtaining a basic lung image, and the method specifically comprises the following steps:
acquiring an initial CT lung image of a target object, and performing data format conversion processing on the initial CT lung image to obtain a lung image with a uniform format;
and carrying out image smoothing and noise reduction on the lung images with the uniform format to obtain high-quality basic lung images.
Fig. 2 shows a flow chart of an initial health simulation lung image acquisition of the present invention.
According to the embodiment of the invention, the lung health simulation model based on deep learning is set up, and the basic lung image is subjected to image key data extraction and image data simulation to obtain an initial health simulation lung image, which specifically comprises the following steps:
s202, carrying out image contour sharpening on the basic lung image to obtain a sharpened basic lung image;
s204, carrying out image contour feature extraction on the sharpened basic lung image to obtain lung size information and contour information;
s206, carrying out image data simulation analysis on the lung size and contour information, and generating an initial health simulation lung image.
It should be noted that the sharpening process can enhance the lung contour characteristics of the basic lung image, so as to obtain more accurate lung size information and contour information. The key data are lung size information and contour information.
According to an embodiment of the present invention, the generating an initial health simulated lung image further comprises:
constructing a lung health simulation model through a TensorFlow framework based on deep learning;
acquiring healthy lung image data and corresponding lung size and contour information data from a pneumonia case image database and big data;
the image data and the information data are used as training samples and are led into a lung health simulation model to carry out image training and result prediction evaluation based on deep learning, and the lung health simulation model with high prediction precision is obtained;
importing the lung size and contour information into a lung health simulation model to perform image simulation for multiple times to obtain a group of simulated lung images;
and calculating the definition of each image in a group of simulated lung images, and selecting the simulated lung image with the highest definition as an initial healthy simulated lung image.
According to the embodiment of the invention, the method for importing the initial health simulation lung image and the pneumonia lesion process parameters into the pneumonia change model for lesion simulation to obtain a lesion simulation image set comprises the following steps:
acquiring an existing historical pneumonia lesion image set from a pneumonia lesion image database;
splitting image data of the historical pneumonia lesion image set according to a preset sequence to obtain a plurality of pneumonia lesion image subsets;
and importing the plurality of pneumonia lesion image subsets into a pneumonia change model for image change analysis, and obtaining pneumonia lesion process parameters according to image changes among different pneumonia lesion image subsets.
It should be noted that the preset sequence is a time sequence of the pneumonia lesion images, and the pneumonia lesion image subset includes one or more images. In the pneumonia lesion process parameter obtaining step, the pneumonia change model analyzes corresponding main lesion image characteristics according to the plurality of pneumonia lesion image subsets, and performs data comparison and analysis on different lesion image characteristics to obtain the pneumonia lesion process parameter. The main lesion image features comprise color depth variation features in the image, image texture variation features, irregular object variation features in the image and the like.
In addition, the historical pneumonia lesion image set is subjected to image data splitting according to a preset sequence to obtain a plurality of pneumonia lesion image subsets, wherein the pneumonia lesion image subsets in the front, middle and rear stages are generally obtained according to a time sequence, namely three pneumonia lesion image subsets.
According to the embodiment of the invention, the obtaining of the existing historical pneumonia lesion image set specifically comprises:
acquiring basic condition information of a historical pneumonia target object;
acquiring basic condition information of a current target object, performing data comparison analysis on the basic condition information of the target object and the basic condition information of each historical target object, and acquiring a plurality of basic condition similarities;
judging whether the similarity of the basic conditions is greater than a preset threshold value, if so, performing similar marking on the corresponding historical target object;
and counting historical target objects with similar marks, and summarizing the pneumonia lesion image sets corresponding to the historical target objects to obtain a historical pneumonia lesion image set.
According to an embodiment of the present invention, the obtaining the plurality of pneumonia lesion image subsets further includes:
acquiring a pneumonia lesion image subset, and performing data splitting on the image subset according to a time sequence to obtain a plurality of pneumonia lesion images;
performing image feature and comparison on two adjacent pneumonia lesion images according to a time sequence to obtain similarity between the images;
and if the similarity is greater than a preset similarity threshold, merging two adjacent pneumonia lesion images, and performing data arrangement on the merged pneumonia lesion images to obtain a pneumonia lesion image subset after image merging.
It should be noted that, in a pneumonia lesion image subset, a plurality of pneumonia lesion images are generally included, and in some cases where a lesion process is not obvious, the change of the obtained general pneumonia lesion images is also not obvious, the similarity of the images is high, and the image characteristic values are also relatively consistent, so that combining the pneumonia lesion images with high similarity at this time can reduce repeated similar images, reduce data redundancy, and thus improve the efficiency of subsequent image analysis and identification.
According to the embodiment of the present invention, the introducing the initial health simulation lung image and the pneumonia lesion process parameters into the pneumonia change model for lesion simulation to obtain a lesion simulation image set further includes:
acquiring basic condition information of a target object, and performing data analysis processing on the basic condition information of the target object to obtain basic data of the target object;
performing pneumonia lesion influence analysis on the target object basic data to obtain a plurality of lesion influence indexes, and calculating an average lesion influence index;
and analyzing to obtain the pneumonia lesion process parameter correction information according to the average lesion influence index.
The target object basic condition information includes information of multiple dimensions such as sex, age, tobacco age, basic disease, and living or working air environment of the target object, and the information is a main factor affecting pneumonia lesions. The target object basic data is a set of specific numerical values of target object basic condition information, and the multiple lesion influence indexes are multiple influence indexes obtained according to different dimension information of the target object.
Fig. 3 shows a lesion simulation image set acquisition flow chart of the present invention.
According to the embodiment of the present invention, the introducing the initial health simulation lung image and the pneumonia lesion process parameters into the pneumonia change model for lesion simulation to obtain a lesion simulation image set further includes:
s302, correcting the pneumonia lesion process parameters according to the lesion process parameter correction information;
s304, importing the initial health simulation lung image and the corrected pneumonia lesion process parameters into a pneumonia change model to simulate a lesion process, and obtaining lesion simulation image data;
s306, dividing the lesion simulation image data into image data according to the image change degree to obtain a plurality of lesion simulation image subsets;
and S308, performing data sorting on the plurality of lesion simulation image subsets to obtain a lesion simulation image set.
It should be noted that the lesion process parameter correction information is correction information obtained according to the target object basic condition information, and the lesion simulation process can be adjusted for different target objects through the parameter correction information, so as to obtain a targeted lesion simulation image set. The lesion simulation image data is all image data of a lung lesion simulation process. The lesion simulation image data is divided into a plurality of lesion simulation image subsets according to the image change degree, the lesion simulation image subsets are generally divided into three lesion simulation image subsets, the three lesion simulation image subsets correspond to the three lesion stages before, during and after the pneumonia lesion process, and the image change characteristics corresponding to different lesion stages are different. The pneumonia change model is an image change model based on deep learning, and the model can calculate and analyze image data according to a given initial image and process change parameters and obtain a changed image result.
According to the embodiment of the present invention, the obtaining of the lesion simulation image set further includes:
acquiring a lesion simulation image set;
comparing and analyzing the target object basic lung image with the lesion simulation image set to obtain a lesion simulation image with the highest similarity to the target object basic lung image in the lesion simulation image set, and marking the lesion simulation image to obtain a marked simulation image;
obtaining the current pneumonia lesion period of the target object according to the marking simulation image and the lesion simulation image set;
acquiring age, occupation, tobacco age, basic diseases and basic information of a living environment condition of a target object;
and generating a targeted medical order scheme of the target object according to the basic information and the pneumonia lesion period of the target object.
It should be noted that, the current pneumonia lesion period of the target object is obtained according to the marker simulation image and the lesion simulation image set, specifically, a corresponding lesion simulation image subset in the lesion simulation image set is found according to the marker simulation image, so as to obtain the pneumonia lesion period, where the number of lesion simulation image subsets is generally three, and corresponds to the front, middle and rear three lesion periods in the pneumonia lesion process. In the specific medical advice scheme for generating the target object, attention items of the target object are comprehensively analyzed according to the pneumonia lesion period and the basic information of the target object, so that a corresponding medical advice scheme is obtained. For example, if the target subject has a number of years of smoking from the target subject basic condition information and the target subject has a pneumonia lesion in the middle and later stages, the target subject is recommended to reduce smoking or stop smoking.
According to the embodiment of the invention, the method further comprises the following steps:
acquiring basic information of age, occupation, tobacco age, basic diseases and living environment conditions of a target object, and evaluating according to the basic information to obtain three evaluation indexes of the basic diseases, environmental influences and health conditions of the target object;
obtaining the prediction probability of various pneumonia complications according to the evaluation index and by combining with historical clinical data;
judging whether the prediction probability is greater than a preset probability threshold value or not, and if so, marking the corresponding pneumonia complications;
and summarizing the marked pneumonia complications to obtain high-probability complications, and generating a corresponding target object pre-care scheme according to the high-probability complications.
It should be noted that. The pneumonia complications include sputum obstruction, pressure sores, respiratory failure, viral myocarditis, arrhythmia, heart failure, septic shock, etc., and the predicted probability of the complications is related to the underlying disease of the target subject, e.g., if the target subject has a low blood pressure underlying cardiovascular disease, the greater the corresponding predicted probability of septic shock in the target subject. In the generating of the corresponding pre-care plan of the target subject, the probability of the occurrence of the complication can be reduced and the symptom of the patient when the complication occurs can be relieved by generating the pre-care plan in advance, for example, if the predicted probability of the respiratory failure of the target subject is high, the corresponding pre-care plan of the target subject includes monitoring of vital signs and maintaining a relatively comfortable air environment of the target subject.
Fig. 4 shows a block diagram of a pneumonia lesion simulation system based on deep learning of the present invention.
The second aspect of the present invention also provides a pneumonia lesion simulation system 4 based on deep learning, which includes: a memory 41 and a processor 42, wherein the memory includes a deep learning-based pneumonia lesion simulation method program, and when the processor executes the deep learning-based pneumonia lesion simulation method program, the method includes the following steps:
acquiring an initial CT lung image, and performing image preprocessing on the initial CT lung image to obtain a basic lung image;
building a lung health simulation model based on deep learning, and performing image key data extraction and image data simulation on a basic lung image to obtain an initial health simulation lung image;
importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model for lesion simulation to obtain a lesion simulation image set;
and collecting the lesion simulation image into an image for splitting, and displaying in a preset terminal device according to a preset mode.
It should be noted that the preset mode is to send the images in the lesion simulation image set to a preset terminal device in a time sequence, where the preset terminal device includes a mobile terminal device and a computer terminal device.
According to the embodiment of the invention, the method for obtaining the initial CT lung image comprises the following steps of:
acquiring an initial CT lung image of a target object, and performing data format conversion processing on the initial CT lung image to obtain a lung image with a uniform format;
and carrying out image smoothing and noise reduction on the lung images with the uniform format to obtain high-quality basic lung images.
According to the embodiment of the invention, the lung health simulation model based on deep learning is established, and the basic lung image is subjected to image key data extraction and image data simulation to obtain an initial health simulation lung image, which specifically comprises the following steps:
carrying out image contour sharpening on the basic lung image to obtain a sharpened basic lung image;
carrying out image contour feature extraction on the sharpened basic lung image to obtain lung size information and contour information;
and carrying out image data simulation analysis on the lung size and contour information, and generating an initial health simulation lung image.
It should be noted that the sharpening process can enhance the lung contour characteristics of the basic lung image, so as to obtain more accurate lung size information and contour information. The key data are lung size information and contour information.
According to the embodiment of the invention, the method for importing the initial health simulation lung image and the pneumonia lesion process parameters into the pneumonia change model for lesion simulation to obtain a lesion simulation image set comprises the following steps:
acquiring an existing historical pneumonia lesion image set from a pneumonia lesion image database;
splitting image data of the historical pneumonia lesion image set according to a preset sequence to obtain a plurality of pneumonia lesion image subsets;
and importing the plurality of pneumonia lesion image subsets into a pneumonia change model for image change analysis, and obtaining pneumonia lesion process parameters according to image changes among different pneumonia lesion image subsets.
It should be noted that the preset sequence is a time sequence of the pneumonia lesion images, and the pneumonia lesion image subset includes one or more images. In the pneumonia lesion process parameter obtaining step, the pneumonia change model analyzes corresponding main lesion image characteristics according to the plurality of pneumonia lesion image subsets, and performs data comparison and analysis on different lesion image characteristics to obtain the pneumonia lesion process parameter. The main lesion image features comprise color depth variation features in the image, image texture variation features, irregular object variation features in the image and the like.
In addition, the historical pneumonia lesion image set is subjected to image data splitting according to a preset sequence to obtain a plurality of pneumonia lesion image subsets, wherein the pneumonia lesion image subsets in the front, middle and rear stages are generally obtained according to a time sequence, namely three pneumonia lesion image subsets.
According to the embodiment of the invention, the obtaining of the existing historical pneumonia lesion image set specifically comprises:
acquiring basic condition information of a historical pneumonia target object;
acquiring basic condition information of a current target object, performing data comparison analysis on the basic condition information of the target object and the basic condition information of each historical target object, and acquiring a plurality of basic condition similarities;
judging whether the similarity of the basic conditions is greater than a preset threshold value, if so, performing similar marking on the corresponding historical target object;
counting historical target objects with similar marks, and summarizing data of the pneumonia lesion image set corresponding to the historical target objects to obtain a historical pneumonia lesion image set.
According to an embodiment of the present invention, the obtaining the plurality of pneumonia lesion image subsets further includes:
acquiring a pneumonia lesion image subset, and performing data splitting on the image subset according to a time sequence to obtain a plurality of pneumonia lesion images;
performing image feature and contrast on two adjacent pneumonia lesion images according to a time sequence to obtain similarity between the images;
and if the similarity is greater than a preset similarity threshold, merging two adjacent pneumonia lesion images, and performing data arrangement on the merged pneumonia lesion images to obtain a pneumonia lesion image subset after image merging.
It should be noted that, in a pneumonia lesion image subset, a plurality of pneumonia lesion images are generally included, and in some cases where lesion processes are not obvious, the obtained plurality of pneumonia lesion images are generally not obvious in image change, the similarity of the images is high, and the image feature values are also relatively consistent, so that the pneumonia lesion images with high similarity are merged at this time, so that repeated similar images can be reduced, the data redundancy is reduced, and the efficiency of subsequent image analysis and identification is improved.
According to the embodiment of the present invention, the introducing the initial health simulation lung image and the pneumonia lesion process parameters into the pneumonia change model for lesion simulation to obtain a lesion simulation image set further includes:
acquiring basic condition information of a target object, and performing data analysis processing on the basic condition information of the target object to obtain basic data of the target object;
performing pneumonia lesion influence analysis on the target object basic data to obtain a plurality of lesion influence indexes, and calculating an average lesion influence index;
and analyzing to obtain the pneumonia lesion process parameter correction information according to the average lesion influence index.
The target object basic condition information includes information of multiple dimensions such as sex, age, tobacco age, and living or working air environment of the target object, and the information is a main factor affecting pneumonia lesions. The target object basic data is a set of specific numerical values of target object basic condition information, and the multiple lesion influence indexes are multiple influence indexes obtained according to different dimension information of the target object.
According to the embodiment of the present invention, the introducing the initial health simulation lung image and the pneumonia lesion process parameters into the pneumonia change model for lesion simulation to obtain a lesion simulation image set further includes:
correcting the pneumonia lesion process parameters according to the lesion process parameter correction information;
importing the initial health simulation lung image and the corrected pneumonia lesion process parameters into a pneumonia change model for lesion process simulation to obtain lesion simulation image data;
dividing the lesion simulation image data into image data according to the image change degree to obtain a plurality of lesion simulation image subsets;
and performing data sorting on the plurality of lesion simulation image subsets to obtain a lesion simulation image set.
It should be noted that the lesion process parameter correction information is correction information obtained according to the target object basic condition information, and the lesion simulation process can be adjusted for different target objects through the parameter correction information, so as to obtain a targeted lesion simulation image set. The lesion simulation image data is all image data of a lung lesion simulation process. The lesion simulation image data is divided into a plurality of lesion simulation image subsets according to the image change degree, the lesion simulation image subsets are generally divided into three lesion simulation image subsets, the three lesion simulation image subsets correspond to the three lesion stages before, during and after the pneumonia lesion process, and the image change characteristics corresponding to different lesion stages are different. The pneumonia change model is an image change model based on deep learning, and the model can calculate and analyze image data according to a given initial image and process change parameters and obtain a changed image result.
The invention discloses a pneumonia lesion simulation method and system based on deep learning. The method comprises the steps of obtaining basic CT lung images of a patient, carrying out health image simulation processing on the lung images to obtain initial health simulation lung images, obtaining historical lung pathological change image sets from a lung pneumonopathy image database by constructing the lung pneumonopathy image database, and leading the image sets into a deep learning-based pneumonia change model for image training to obtain pneumonia pathological change process parameters. And importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model to perform lesion simulation to obtain a lesion simulation image set, and performing data comparison and analysis on the patient real CT lung image and the lesion simulation image set to obtain pneumonia condition reference information so as to assist a doctor in judging the pneumonia condition of the patient. The invention can assist doctors in comprehensively analyzing the lung images of patients, improve the working efficiency of doctors and reduce the misdiagnosis rate.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or in other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a ROM, a RAM, a magnetic or optical disk, or various other media that can store program code.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (10)
1. A pneumonia lesion simulation method based on deep learning is characterized by comprising the following steps:
acquiring an initial CT lung image, and performing image preprocessing on the initial CT lung image to obtain a basic lung image;
building a lung health simulation model based on deep learning, and performing image key data extraction and image data simulation on a basic lung image to obtain an initial health simulation lung image;
importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model for lesion simulation to obtain a lesion simulation image set;
collecting the lesion simulation image into an image for splitting, and displaying in a preset terminal device according to a preset mode;
wherein, the obtaining of the lesion simulation image set further comprises:
acquiring a lesion simulation image set;
comparing and analyzing the target object basic lung image with the lesion simulation image set to obtain a lesion simulation image with the highest similarity to the target object basic lung image in the lesion simulation image set, and marking the lesion simulation image to obtain a marked simulation image;
obtaining the current pneumonia lesion period of the target object according to the marking simulation image and the lesion simulation image set;
acquiring age, occupation, tobacco age, basic diseases and basic information of a living environment condition of a target object;
and generating a targeted medical order scheme of the target object according to the basic information and the pneumonia lesion period of the target object.
2. The pneumonia lesion simulation method based on deep learning of claim 1, wherein the initial CT lung image is obtained, and the initial CT lung image is subjected to image preprocessing to obtain a basic lung image, specifically:
acquiring an initial CT lung image of a target object, and performing data format conversion processing on the initial CT lung image to obtain a lung image with a uniform format;
and carrying out image smoothing and noise reduction on the lung images with the uniform format to obtain high-quality basic lung images.
3. The pneumonia lesion simulation method based on deep learning of claim 1, wherein the lung health simulation model based on deep learning is set up, and the basic lung image is subjected to image key data extraction and image data simulation to obtain an initial health simulation lung image, and specifically: carrying out image contour sharpening on the basic lung image to obtain a sharpened basic lung image;
carrying out image contour feature extraction on the sharpened basic lung image to obtain lung size information and contour information;
and carrying out image data simulation analysis on the lung size and contour information, and generating an initial health simulation lung image.
4. The method for simulating pneumonia lesion based on deep learning of claim 1, wherein the introducing the initial healthy simulated lung image and pneumonia lesion process parameters into pneumonia change model for lesion simulation to obtain lesion simulation image set includes:
acquiring an existing historical pneumonia lesion image set from a pneumonia lesion image database;
splitting image data of the historical pneumonia lesion image set according to a preset sequence to obtain a plurality of pneumonia lesion image subsets;
and importing the plurality of pneumonia lesion image subsets into a pneumonia change model for image change analysis, and obtaining pneumonia lesion process parameters according to image changes among different pneumonia lesion image subsets.
5. The method for simulating pneumonia lesion based on deep learning of claim 1, wherein the introducing the initial healthy simulated lung image and pneumonia lesion process parameters into a pneumonia change model for lesion simulation to obtain a lesion simulation image set further comprises:
acquiring basic condition information of a target object, and performing data analysis processing on the basic condition information of the target object to obtain basic data of the target object;
performing pneumonia lesion influence analysis on the target object basic data to obtain a plurality of lesion influence indexes, and calculating an average lesion influence index;
and analyzing to obtain the pneumonia lesion process parameter correction information according to the average lesion influence index.
6. The pneumonia lesion simulation method based on deep learning of claim 1 wherein the introducing of the initial healthy simulated lung image and the pneumonia lesion process parameters into a pneumonia change model for lesion simulation results in a lesion simulated image set, further comprising:
correcting the pneumonia lesion process parameters according to the lesion process parameter correction information;
importing the initial health simulation lung image and the corrected pneumonia lesion process parameters into a pneumonia change model for lesion process simulation to obtain lesion simulation image data;
dividing the lesion simulation image data into image data according to the image change degree to obtain a plurality of lesion simulation image subsets;
and performing data sorting on the plurality of lesion simulation image subsets to obtain a lesion simulation image set.
7. A pneumonia lesion simulation system based on deep learning, the system comprising: the pneumonia lesion simulation method program based on deep learning is executed by the processor, and the pneumonia lesion simulation method program based on deep learning realizes the following steps:
acquiring an initial CT lung image, and performing image preprocessing on the initial CT lung image to obtain a basic lung image;
building a lung health simulation model based on deep learning, and performing image key data extraction and image data simulation on a basic lung image to obtain an initial health simulation lung image;
importing the initial health simulation lung image and pneumonia lesion process parameters into a pneumonia change model for lesion simulation to obtain a lesion simulation image set;
and collecting the lesion simulation image into an image for splitting, and displaying in a preset terminal device according to a preset mode.
8. The system of claim 7, wherein the system for acquiring the initial CT lung image and performing image preprocessing on the initial CT lung image to obtain a basic lung image is specifically:
acquiring an initial CT lung image of a target object, and performing data format conversion processing on the initial CT lung image to obtain a lung image with a uniform format;
and carrying out image smoothing and noise reduction on the lung images with the uniform format to obtain high-quality basic lung images.
9. The system of claim 7, wherein the initial healthy lung simulation image and the pneumonia lesion process parameters are introduced into a pneumonia change model for lesion simulation to obtain a lesion simulation image set, and further comprising:
acquiring basic condition information of a target object, and performing data analysis processing on the basic condition information of the target object to obtain basic data of the target object;
performing pneumonia lesion influence analysis on the target object basic data to obtain a plurality of lesion influence indexes, and calculating an average lesion influence index;
and analyzing to obtain the pneumonia lesion process parameter correction information according to the average lesion influence index.
10. The system of claim 7, wherein the initial healthy lung simulation image and the pneumonia lesion process parameters are introduced into a pneumonia change model for lesion simulation to obtain a lesion simulation image set, and further comprising:
correcting the pneumonia lesion process parameters according to the lesion process parameter correction information;
importing the initial health simulation lung image and the corrected pneumonia lesion process parameters into a pneumonia change model for lesion process simulation to obtain lesion simulation image data;
dividing the lesion simulation image data into image data according to the image change degree to obtain a plurality of lesion simulation image subsets;
and performing data sorting on the plurality of lesion simulation image subsets to obtain a lesion simulation image set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211012475.3A CN115115620B (en) | 2022-08-23 | 2022-08-23 | Pneumonia lesion simulation method and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211012475.3A CN115115620B (en) | 2022-08-23 | 2022-08-23 | Pneumonia lesion simulation method and system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115115620A true CN115115620A (en) | 2022-09-27 |
| CN115115620B CN115115620B (en) | 2022-12-13 |
Family
ID=83335947
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211012475.3A Active CN115115620B (en) | 2022-08-23 | 2022-08-23 | Pneumonia lesion simulation method and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115115620B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116051476A (en) * | 2022-12-23 | 2023-05-02 | 广州市番禺区中心医院 | Automatic evaluation system for pneumosepsis based on scanning image analysis |
| CN116739890A (en) * | 2023-06-26 | 2023-09-12 | 强联智创（北京）科技有限公司 | Method and equipment for training generation model for generating healthy blood vessel image |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105640577A (en) * | 2015-12-16 | 2016-06-08 | 深圳市智影医疗科技有限公司 | Method and system automatically detecting local lesion in radiographic image |
| KR20180040287A (en) * | 2016-10-12 | 2018-04-20 | (주)헬스허브 | System for interpreting medical images through machine learnings |
| US20180137941A1 (en) * | 2015-06-02 | 2018-05-17 | Infervision Co., Ltd. | Method For Analysing Medical Treatment Data Based On Deep Learning and Intelligence Analyser Thereof |
| CN108877922A (en) * | 2017-05-15 | 2018-11-23 | 沈渊瑶 | Lesion degree judges system and method |
| CN109637669A (en) * | 2018-11-22 | 2019-04-16 | 中山大学 | Generation method, device and the storage medium of therapeutic scheme based on deep learning |
| JP2020018705A (en) * | 2018-08-02 | 2020-02-06 | キヤノンメディカルシステムズ株式会社 | Medical image processing device, image formation method and image formation program |
| US20200118691A1 (en) * | 2018-10-10 | 2020-04-16 | Lukasz R. Kiljanek | Generation of Simulated Patient Data for Training Predicted Medical Outcome Analysis Engine |
| CN111403042A (en) * | 2020-06-08 | 2020-07-10 | 成都泰盟软件有限公司 | AI-based virtual standard patient simulation system |
| CN111738997A (en) * | 2020-06-11 | 2020-10-02 | 浙江大学 | Method for calculating new coronary pneumonia lesion area ratio based on deep learning |
| CN111862075A (en) * | 2020-07-30 | 2020-10-30 | 西南医科大学 | Lung image analysis system and method based on deep learning |
| CN112258511A (en) * | 2020-09-08 | 2021-01-22 | 广东省智能制造研究所 | Partial differential equation-based patient lung lesion prediction method and system |
| CN112690815A (en) * | 2020-11-30 | 2021-04-23 | 四川大学 | System and method for assisting in diagnosing lesion grade based on lung image report |
| CN112820398A (en) * | 2021-01-22 | 2021-05-18 | 神威超算(北京)科技有限公司 | Artificial intelligence-based lung diagnosis auxiliary system and method |
| CN215868349U (en) * | 2021-08-09 | 2022-02-18 | 漳州卫生职业学院 | Lung palpation anthropomorphic dummy |
-
2022
- 2022-08-23 CN CN202211012475.3A patent/CN115115620B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180137941A1 (en) * | 2015-06-02 | 2018-05-17 | Infervision Co., Ltd. | Method For Analysing Medical Treatment Data Based On Deep Learning and Intelligence Analyser Thereof |
| CN105640577A (en) * | 2015-12-16 | 2016-06-08 | 深圳市智影医疗科技有限公司 | Method and system automatically detecting local lesion in radiographic image |
| KR20180040287A (en) * | 2016-10-12 | 2018-04-20 | (주)헬스허브 | System for interpreting medical images through machine learnings |
| CN108877922A (en) * | 2017-05-15 | 2018-11-23 | 沈渊瑶 | Lesion degree judges system and method |
| JP2020018705A (en) * | 2018-08-02 | 2020-02-06 | キヤノンメディカルシステムズ株式会社 | Medical image processing device, image formation method and image formation program |
| US20200118691A1 (en) * | 2018-10-10 | 2020-04-16 | Lukasz R. Kiljanek | Generation of Simulated Patient Data for Training Predicted Medical Outcome Analysis Engine |
| CN109637669A (en) * | 2018-11-22 | 2019-04-16 | 中山大学 | Generation method, device and the storage medium of therapeutic scheme based on deep learning |
| CN111403042A (en) * | 2020-06-08 | 2020-07-10 | 成都泰盟软件有限公司 | AI-based virtual standard patient simulation system |
| CN111738997A (en) * | 2020-06-11 | 2020-10-02 | 浙江大学 | Method for calculating new coronary pneumonia lesion area ratio based on deep learning |
| CN111862075A (en) * | 2020-07-30 | 2020-10-30 | 西南医科大学 | Lung image analysis system and method based on deep learning |
| CN112258511A (en) * | 2020-09-08 | 2021-01-22 | 广东省智能制造研究所 | Partial differential equation-based patient lung lesion prediction method and system |
| CN112690815A (en) * | 2020-11-30 | 2021-04-23 | 四川大学 | System and method for assisting in diagnosing lesion grade based on lung image report |
| CN112820398A (en) * | 2021-01-22 | 2021-05-18 | 神威超算(北京)科技有限公司 | Artificial intelligence-based lung diagnosis auxiliary system and method |
| CN215868349U (en) * | 2021-08-09 | 2022-02-18 | 漳州卫生职业学院 | Lung palpation anthropomorphic dummy |
Non-Patent Citations (4)
| Title |
|---|
| KANG ZHANG 等: "Clinically Applicable AI System for Accurate Diagnosis, Quantitative Measurements, and Prognosis of COVID-19 Pneumonia Using Computed Tomography", 《CELL》 * |
| XINXU WEI 等: "Deep Pneumonia: Attention-BLearning for Class-Imbalanced Pneumonia Lesion ased Contrastive Recognition in Chest X-rays", 《ARXIV》 * |
| 李芳芳 等: "基于注意力机制的肺炎辅助诊断模型", 《兰州文理学院学报（自然科学版）》 * |
| 赵小二 等: "不同阶段新型冠状病毒肺炎的CT影像演变分析", 《实用放射学杂志》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116051476A (en) * | 2022-12-23 | 2023-05-02 | 广州市番禺区中心医院 | Automatic evaluation system for pneumosepsis based on scanning image analysis |
| CN116051476B (en) * | 2022-12-23 | 2023-08-18 | 广州市番禺区中心医院 | Automatic evaluation system for pneumosepsis based on scanning image analysis |
| CN116739890A (en) * | 2023-06-26 | 2023-09-12 | 强联智创（北京）科技有限公司 | Method and equipment for training generation model for generating healthy blood vessel image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115115620B (en) | 2022-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115115620B (en) | Pneumonia lesion simulation method and system based on deep learning | |
| CN110969626B (en) | Method for extracting hippocampus of human brain nuclear magnetic resonance image based on 3D neural network | |
| CN108109700B (en) | Method and device for evaluating curative effect of chronic disease | |
| EP3306500A1 (en) | Method for analysing medical treatment data based on deep learning, and intelligent analyser thereof | |
| CN108937912B (en) | Automatic arrhythmia analysis method based on deep neural network | |
| CN113095382B (en) | Interpretable tuberculosis classification network identification method based on CT image | |
| CN111243753B (en) | Multi-factor correlation interactive analysis method for medical data | |
| CN111738997A (en) | Method for calculating new coronary pneumonia lesion area ratio based on deep learning | |
| CN111956214A (en) | QRS wave automatic detection method based on U-net end-to-end neural network model | |
| CN113288157A (en) | Arrhythmia classification method based on depth separable convolution and improved loss function | |
| CN116597985A (en) | Survival rate prediction model training method, survival period prediction method, survival rate prediction device and survival rate prediction equipment | |
| CN109410074A (en) | Intelligent core protects method and system | |
| CN111080592A (en) | Rib extraction method and device based on deep learning | |
| CN111261283B (en) | Electrocardiosignal deep neural network modeling method based on pyramid convolution layer | |
| CN111160812B (en) | Diagnostic information evaluation method, display method, and storage medium | |
| CN116246779B (en) | Dental diagnosis and treatment scheme generation method and system based on user image data | |
| CN114708264B (en) | Light spot quality judging method, device, equipment and storage medium | |
| CN116469570A (en) | Malignant tumor complication analysis method based on electronic medical record | |
| CN116504392A (en) | Intelligent auxiliary diagnosis prompt system based on data analysis | |
| CN113011510B (en) | Bronchial classification and model training method and device and electronic equipment | |
| CN112561935B (en) | Intelligent classification method, device and equipment for brain images | |
| CN114093521A (en) | Random forest based method and system for estimating blood sugar by reconstructing homogenized samples | |
| CN113706644A (en) | Image processing method, image processing apparatus, and storage medium | |
| CN112364924A (en) | Deep learning-based oral medical image identification method | |
| CN112101021A (en) | Method, device and equipment for realizing standard word mapping |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |